No description available.

Automated Face Recognition is the process of determining a subject's identity from digital imagery of their face without user intervention. The term in fact encompasses two distinct tasks; Face Verficiation is the process of verifying a subject's claimed identity while Face Identification involves selecting the most likely identity from a database of subjects. This dissertation focuses on the task of Face Verification, which has a myriad of applications in security ranging from border control to personal banking. Recently the use of 3D facial imagery has found favour in the research community due to its inherent robustness to the pose and illumination variations which plague the 2D modality. The field of 3D face recognition is, however, yet to fully mature and there remain many unanswered research questions particular to the modality. The relative expense and specialty of 3D acquisition devices also means that the availability of databases of 3D face imagery lags significantly behind that of standard 2D face images. Human recognition of faces is rooted in an inherently 2D visual system and much is known regarding the use of 2D image information in the recognition of individuals. The corresponding knowledge of how discriminative information is distributed in the 3D modality is much less well defined. This dissertations addresses these issues through the use of decompositional techniques. Decomposition alleviates the problems associated with dimensionality explosion and the Small Sample Size (SSS) problem and spatial decomposition is a technique which has been widely used in face recognition. The application of decomposition in the frequency domain, however, has not received the same attention in the literature. The use of decomposition techniques allows a map ping of the regions (both spatial and frequency) which contain the discriminative information that enables recognition. In this dissertation these techniques are covered in significant detail, both in terms of practical issues in the respective domains and in terms of the underlying distributions which they expose. Significant discussion is given to the manner in which the inherent information of the human face is manifested in the 2D and 3D domains and how these two modalities inter-relate. This investigation is extended to cover also the manner in which the decomposition techniques presented can be recombined into a single decision. Two new methods for learning the weighting functions for both the sum and product rules are presented and extensive testing against established methods is presented. Knowledge acquired from these examinations is then used to create a combined technique termed Log-Gabor Templates. The proposed technique utilises both the spatial and frequency domains to extract superior performance to either in isolation. Experimentation demonstrates that the spatial and frequency domain decompositions are complimentary and can combined to give improved performance and robustness.

Faire le lien entre la morphologie d'un signal sonore et certains de ses attributs perceptifs est une étape capitale dans l'élaboration d'un synthétiseur proposant un contrôle intuitif. Certains aspects de cette morphologie peuvent être caractérisés au moyen de "descripteurs acoustiques". Lorsqu'ils sont choisis judicieusement, ces descripteurs permettent de classer des signaux dans des catégories ayant un sens perceptif ; ceci permet d'établir un lien entre morphologie et perception. Dans le travail présenté ici, on s'intéresse en particulier à la catégorisation perceptive de sons d'impact.La plupart des descripteurs considérés ici se construisent à partir d'une modélisation paramétrique du signal. Dans notre cas, la modélisation la plus appropriée semble être la décomposition en somme de sinusoïdes amorties. Une estimation stable et rigoureuse des paramètres du modèle étant essentielle au calcul des descripteurs, on se penche sur la comparaison de plusieurs méthodes de décomposition. Il ressort que la méthode à haute résolution ESPRIT semble la plus indiquée, mais qu'elle ne peut pas être utilisée sous sa forme classique. On propose donc différentes adaptations. En particulier, on s'intéresse à l'application d'ESPRIT dans des repères de Gabor. En outre, on propose des méthodes pour maximiser le caractère parcimonieux de la décomposition.On étudie finalement un cas d'application concret : à partir d'une banque de sons enregistrés en chambre anéchoïque résultant d'impacts sur divers objets du quotidien, on évalue la pertinence d'un ensemble de descripteurs pour la catégorisation en fonction du matériau perçu. / Linking an audio signal morphology with some of its perceptual attributes is a key step when elaborating a intuitively controlled synthesizer. Some of these morphology aspects can be characterized using "acoustical descriptors". When chosen wisely, descriptors can allow a classification of audio signals in categories which are perceptually relevant ; in such cases, this approach establishes a link between morphology and perception. The present work focuses on the perceptual categorization of impact sounds.Most of the descriptors proposed here are computed using a parametrized description of the signal. Here, the most appropriate parametrization seems to be a decomposition in exponentially damped sinusoids. A robust and stable estimation of the model parameters being essential to the computation of relevant descriptors, different parametrization methods are described and compared. From these comparisons, it appears that the high-resolution method ESPRIT is the most appropriate, but that it cannot be applied in its classical form. Several adaptations are therefore investigated. In particular, the application of ESPRIT in Gabor frames is considered. Besides, a method is proposed in order to minimize the number of components necessary for a satisfactory decomposition.Finally, a concrete application is addressed : from an impact sounds bank recorded in an anechoic chamber, elaborated with a wide range of everyday-life objects, the relevance of several acoustical descriptors for the perceptual categorization of the perceived material is investigated.

Os sistemas de recuperação de imagens por conteúdo (CBIR -Content-based Image Retrieval) possuem a habilidade de retornar imagens utilizando como chave de busca outras imagens. Considerando uma imagem de consulta, o foco de um sistema CBIR é pesquisar no banco de dados as "n" imagens mais similares à imagem de consulta de acordo com um critério dado. Este trabalho de pesquisa foi direcionado na geração de vetores de características para um sistema CBIR considerando bancos de imagens médicas, para propiciar tal tipo de consulta. Um vetor de características é uma representação numérica sucinta de uma imagem ou parte dela, descrevendo seus detalhes mais representativos. O vetor de características é um vetor "n"-dimensional contendo esses valores. Essa nova representação da imagem pode ser armazenada em uma base de dados, e assim, agilizar o processo de recuperação de imagens. Uma abordagem alternativa para caracterizar imagens para um sistema CBIR é a transformação do domínio. A principal vantagem de uma transformação é sua efetiva caracterização das propriedades locais da imagem. Recentemente, pesquisadores das áreas de matemática aplicada e de processamento de sinais desenvolveram técnicas práticas de "wavelet" para a representação multiescala e análise de sinais. Estas novas ferramentas diferenciam-se das tradicionais técnicas de Fourier pela forma de localizar a informação no plano tempo-freqüência; basicamente, elas têm a capacidade de mudar de uma resolução para outra, o que faz delas especialmente adequadas para a análise de sinais não estacionários. A transformada "wavelet" consiste de um conjunto de funções base que representa o sinal em diferentes bandas de freqüência, cada uma com resoluções distintas correspondentes a cada escala. Estas foram aplicadas com sucesso na compressão, melhoria, análise, classificação, caracterização e recuperação de imagens. Uma das áreas beneficiadas, onde essas propriedades têm encontrado grande relevância, é a área médica, através da representação e descrição de imagens médicas. Este trabalho descreve uma abordagem para um banco de imagens médicas, que é orientada à extração de características para um sistema CBIR baseada na decomposição multiresolução de "wavelets" utilizando os filtros de Daubechies e Gabor. Essas novas características de imagens foram também testadas utilizando uma estrutura de indexação métrica "Slim-tree". Assim, pode-se aumentar o alcance semântico do sistema cbPACS (Content-Based Picture Archiving and Comunication Systems), atualmente em desenvolvimento conjunto entre o Grupo de Bases de Dados e Imagens do ICMC--USP e o Centro de Ciências de Imagens e Física Médica do Hospital das Clínicas de Riberão Preto-USP. / Content-based image retrieval (CBIR) refers to the ability to retrieve images on the basis of the image content. Given a query image, the goal of a CBIR system is to search the database and return the "n" most similar (close) ones to the query image according to a given criteria. Our research addresses the generation of feature vectors of a CBIR system for medical image databases. A feature vector is a numeric representation of an image or part of it over its representative aspects. The feature vector is a "n"-dimensional vector organizing such values. This new image representation can be stored into a database and allow a fast image retrieval. An alternative for image characterization for a CBIR system is the domain transform. The principal advantage of a transform is its effective characterization for their local image properties. In the past few years, researches in applied mathematics and signal processing have developed practical "wavelet" methods for the multiscale representation and analysis of signals. These new tools differ from the traditional Fourier techniques by the way in which they localize the information in the time-frequency plane; in particular, they are capable of trading one type of resolution for the other, which makes them especially suitable for the analysis of non-stationary signals. The "wavelet" transform is a set of basis functions that represents signals in different frequency bands, each one with a resolution matching its scale. They have been successfully applied to image compression, enhancements, analysis, classifications, characterization and retrieval. One privileged area of application where these properties have been found to be relevant is medical imaging. In this work we describe an approach to CBIR for medical image databases focused on feature extraction based on multiresolution "wavelets" decomposition, taking advantage of the Daubechies and Gabor. Fundamental to our approach is how images are characterized, such that the retrieval procedure can bring similar images within the domain of interest, using a metric structure indexing, like the "Slim-tree". Thus, it increased the semantic capability of the cbPACS(Content-Based Picture Archiving and Comunication Systems), currently in joined developing between the Database and Image Group of the ICMC--USP and the Science Center for Images and Physical Medic of the Clinics Hospital of Riberão Preto--USP.

[pt] O presente trabalho apresenta um estudo sobre técnicas de construção de um modelo tridimensional de objetos a partir unicamente da informação de textura. Estas técnicas são baseadas na medida da deformação da textura ao longo de uma superfície, obtendo assim a orientação do vetor normal à superfície em cada ponto. De posse da orientação é possível construir um modelo tridimensional do objeto. São avaliados três métodos. O primeiro emprega Filtros de Gabor e momentos de segunda ordem como medida de textura e os outros dois estimam a transformação afim entre recortes de igual tamanho na imagem. A estimativa da transformação afim tem ênfase especial neste trabalho por ser um passo fundamental no algoritmo que mede a deformação da textura. Os métodos foram validados em diferentes etapas, de forma a avaliar: estimativa da transformação afim; decomposição em ângulos; e reconstrução do modelo 3D a partir do mapa de orientação, também conhecido como mapa de agulhas. A avaliação experimental foi realizada com imagens sintéticas e fotos de objetos reais. Os resultados mostram a aplicabilidade, dificuldades e restrições dos métodos analisados. / [en] The current work presents a study about methods for 3D object shape reconstruction based on their texture information. These methods, called Shape from Texture, measure texture deformation along object surface, obtaining the orientation in each point of the image. Having the orientation in each point (a needle map) it is possible to construct the object 3D model. Three methods are studied in this dissertation. One of these methods uses Gabor Filters and second order moments, and other two that estimate the affine transform between images patches. The affine estimation problem gets emphasis in the present work since it is an essential step in most Shape from Texture algorithms. The methods were tested in separate steps: evaluate the affine transform estimation; the decomposition of the affine matrix in slant and tilt angles; and the 3D model reconstruction using the needle map. Both synthetic and real images were used on the experiments. The results clearly show the applicability, difficulties and restrictions of the investigated methods.

No description available.

Σκοπός της παρούσας διπλωματικής εργασίας είναι η κατασκευή ενός συστήματος που να μπορεί να ξεχωρίζει τα οχήματα από άλλα αντικείμενα με τη χρήση κυματιδίου Haar και φίλτρου Gabor (εξαγωγή χαρακτηριστικών) και SVM, RBF για ταξινόμηση. / The aim of this thesis is the construction of a system that will be able to distiguish vehicles from other objects using Haar and Gabor filter (export characteristic) and SVM, RBF for classification.

Ce travail de thèse vise à mettre en place un système d’extraction de réseau routier en milieu urbain à partir d’image satellite à très haute résolution. Dans ce contexte, nous avons proposé deux méthodes de localisation de routes. La première approche est fondée sur la procédure de conversion de l’image vers un format vectoriel. L’originalité de cette approche réside dans l’utilisation d’une méthode géométrique pour assurer le passage vers une représentation vectorielle de l’image d’origine et la mise en place d’un formalisme logique fondé sur un ensemble de critères perceptifs permettant le filtrage de l’information inutile et l’extraction des structures linéaires. Dans la deuxième approche, nous avons proposé un algorithme fondé sur la théorie des ondelettes, il met particulièrement en évidence les deux volets multi-résolution et multi-direction. Nous proposons donc une approche de localisation des routes mettant en jeux l'information fréquentielle multi directionnelle issue de la transformée en ondelette Log-Gabor. Dans l’étape de localisation, nous avons présenté deux détecteurs de routes qui exploitent l’information radiométrique, géométrique et fréquentielle. Cependant, ces informations ne permettent pas un résultat exact et précis. Pour remédier à ce problème, un algorithme de suivi s’avère nécessaire. Nous proposons la reconstruction de réseaux routiers par des courbes NURBS. Cette approche est basée sur un ensemble de points de repères identifiés dans la phase de localisation. Elle propose un nouveau concept, que nous avons désigné par NURBSC, basé sur les contraintes géométriques des formes à approximer. Nous connectons les segments de route identifiés afin d’obtenir des tracés continus propres aux routes. / The aim of this thesis is to establish a road network extraction system in urban areas from very high resolution satellite images. In this context, we proposed two approaches to locate roads. The first one is based on the process of converting the image into a vector form. The originality of this approach lies in the use of a geometric method to ensure the shift into a vector representation of the original image and the establishment of a logical formalism based on a set of perceptual criteria. It allows the filtering of unnecessary information and extracting linear structures. In the second approach, we proposed an algorithm based on the wavelet theory, it particularly highlights the two axis multi-resolution and multi-direction. Thus, we introduce a road localization approach, which manage the frequency multidirectional data resulting from the transform using the Log-Gabor wavelet. In the localization step, we presented two road detectors, which are capable of exploiting the radiometric, geometric and frequency data. However, this data cannot allow accurate and precise results. To overcome this drawback, a tracking algorithm is needed. We propose the reconstruction of road networks by NURBS curves. This approach is based on a landmark set of points identified in the localization phase and presents a new concept, noted by NURBSC. NURBSC is based on the geometrical constraints of shapes to be approximated. We connect road segments identified in order to obtain continuous road network.

Os sistemas de recuperação de imagens por conteúdo (CBIR -Content-based Image Retrieval) possuem a habilidade de retornar imagens utilizando como chave de busca outras imagens. Considerando uma imagem de consulta, o foco de um sistema CBIR é pesquisar no banco de dados as "n" imagens mais similares à imagem de consulta de acordo com um critério dado. Este trabalho de pesquisa foi direcionado na geração de vetores de características para um sistema CBIR considerando bancos de imagens médicas, para propiciar tal tipo de consulta. Um vetor de características é uma representação numérica sucinta de uma imagem ou parte dela, descrevendo seus detalhes mais representativos. O vetor de características é um vetor "n"-dimensional contendo esses valores. Essa nova representação da imagem pode ser armazenada em uma base de dados, e assim, agilizar o processo de recuperação de imagens. Uma abordagem alternativa para caracterizar imagens para um sistema CBIR é a transformação do domínio. A principal vantagem de uma transformação é sua efetiva caracterização das propriedades locais da imagem. Recentemente, pesquisadores das áreas de matemática aplicada e de processamento de sinais desenvolveram técnicas práticas de "wavelet" para a representação multiescala e análise de sinais. Estas novas ferramentas diferenciam-se das tradicionais técnicas de Fourier pela forma de localizar a informação no plano tempo-freqüência; basicamente, elas têm a capacidade de mudar de uma resolução para outra, o que faz delas especialmente adequadas para a análise de sinais não estacionários. A transformada "wavelet" consiste de um conjunto de funções base que representa o sinal em diferentes bandas de freqüência, cada uma com resoluções distintas correspondentes a cada escala. Estas foram aplicadas com sucesso na compressão, melhoria, análise, classificação, caracterização e recuperação de imagens. Uma das áreas beneficiadas, onde essas propriedades têm encontrado grande relevância, é a área médica, através da representação e descrição de imagens médicas. Este trabalho descreve uma abordagem para um banco de imagens médicas, que é orientada à extração de características para um sistema CBIR baseada na decomposição multiresolução de "wavelets" utilizando os filtros de Daubechies e Gabor. Essas novas características de imagens foram também testadas utilizando uma estrutura de indexação métrica "Slim-tree". Assim, pode-se aumentar o alcance semântico do sistema cbPACS (Content-Based Picture Archiving and Comunication Systems), atualmente em desenvolvimento conjunto entre o Grupo de Bases de Dados e Imagens do ICMC--USP e o Centro de Ciências de Imagens e Física Médica do Hospital das Clínicas de Riberão Preto-USP. / Content-based image retrieval (CBIR) refers to the ability to retrieve images on the basis of the image content. Given a query image, the goal of a CBIR system is to search the database and return the "n" most similar (close) ones to the query image according to a given criteria. Our research addresses the generation of feature vectors of a CBIR system for medical image databases. A feature vector is a numeric representation of an image or part of it over its representative aspects. The feature vector is a "n"-dimensional vector organizing such values. This new image representation can be stored into a database and allow a fast image retrieval. An alternative for image characterization for a CBIR system is the domain transform. The principal advantage of a transform is its effective characterization for their local image properties. In the past few years, researches in applied mathematics and signal processing have developed practical "wavelet" methods for the multiscale representation and analysis of signals. These new tools differ from the traditional Fourier techniques by the way in which they localize the information in the time-frequency plane; in particular, they are capable of trading one type of resolution for the other, which makes them especially suitable for the analysis of non-stationary signals. The "wavelet" transform is a set of basis functions that represents signals in different frequency bands, each one with a resolution matching its scale. They have been successfully applied to image compression, enhancements, analysis, classifications, characterization and retrieval. One privileged area of application where these properties have been found to be relevant is medical imaging. In this work we describe an approach to CBIR for medical image databases focused on feature extraction based on multiresolution "wavelets" decomposition, taking advantage of the Daubechies and Gabor. Fundamental to our approach is how images are characterized, such that the retrieval procedure can bring similar images within the domain of interest, using a metric structure indexing, like the "Slim-tree". Thus, it increased the semantic capability of the cbPACS(Content-Based Picture Archiving and Comunication Systems), currently in joined developing between the Database and Image Group of the ICMC--USP and the Science Center for Images and Physical Medic of the Clinics Hospital of Riberão Preto--USP.

La séparation aveugle de source consiste à estimer les signaux de sources uniquement à partir des mélanges observés. Le problème peut être séparé en deux catégories en fonction du modèle de mélange: mélanges instantanés, où le retard et la réverbération (effet multi-chemin) ne sont pas pris en compte, et des mélanges convolutives qui sont plus généraux mais plus compliqués. De plus, le bruit additif au niveaux des capteurs et le réglage sous-déterminé, où il y a moins de capteurs que les sources, rendent le problème encore plus difficile.Dans cette thèse, tout d'abord, nous avons étudié le lien entre deux méthodes existantes pour les mélanges instantanés: analyse des composants indépendants (ICA) et analyse des composant parcimonieux (SCA). Nous avons ensuite proposé une nouveau formulation qui fonctionne dans les cas déterminés et sous-déterminés, avec et sans bruit. Les évaluations numériques montrent l'avantage des approches proposées.Deuxièmement, la formulation proposés est généralisés pour les mélanges convolutifs avec des signaux de parole. En intégrant un nouveau modèle d'approximation, les algorithmes proposés fonctionnent mieux que les méthodes existantes, en particulier dans des scénarios bruyant et / ou de forte réverbération.Ensuite, on prend en compte la technique de décomposition morphologique et l'utilisation de parcimonie structurée qui conduit à des algorithmes qui peuvent mieux exploiter les structures des signaux audio. De telles approches sont testées pour des mélanges convolutifs sous-déterminés dans un scénario non-aveugle.Enfin, en bénéficiant du modèle NMF (factorisation en matrice non-négative), nous avons combiné l'hypothèse de faible-rang et de parcimonie et proposé de nouvelles approches pour les mélanges convolutifs sous-déterminés. Les expériences illustrent la bonne performance des algorithmes proposés pour les signaux de musique, en particulier dans des scénarios de forte réverbération. / Blind source separation (BSS) consists of estimating the source signals only from the observed mixtures. The problem can be divided into two categories according to the mixing model: instantaneous mixtures, where delay and reverberation (multi-path effect) are not taken into account, and convolutive mixtures which are more general but more complicated. Moreover, the additive noise at the sensor level and the underdetermined setting, where there are fewer sensors than the sources, make the problem even more difficult.In this thesis, we first studied the link between two existing methods for instantaneous mixtures: independent component analysis (ICA) and sparse component analysis (SCA). We then proposed a new formulation that works in both determined and underdetermined cases, with and without noise. Numerical evaluations show the advantage of the proposed approaches.Secondly, the proposed formulation is generalized for convolutive mixtures with speech signals. By integrating a new approximation model, the proposed algorithms work better than existing methods, especially in noisy and/or high reverberation scenarios.Then, we take into account the technique of morphological decomposition and the use of structured sparsity which leads to algorithms that can better exploit the structures of audio signals. Such approaches are tested for underdetermined convolutive mixtures in a non-blind scenario.At last, being benefited from the NMF model, we combined the low-rank and sparsity assumption and proposed new approaches for under-determined convolutive mixtures. The experiments illustrate the good performance of the proposed algorithms for music signals, especially in strong reverberation scenarios.